Anthropic vs The Pentagon: Who Wins? | Cursor Hits $2BN in ARR | Block's 40% Headcount Reduction

1. 0:00 – 1:16

   Intro

   1. RORory O’Driscoll0:00

      Whatever it is that's in the water at Anthropic, it's working and it has created unity. You've got to believe that the next round for this, Anthropic and SpaceX, are all public offerings. Tesla trades at a trillion today. I think if Elon died tomorrow they'd trade at $200BN. OpenAI trades at $800BN today. I think if Sam Altman died tomorrow morning, we'd trade at $600BN.

   2. HSHarry Stebbings0:18

      This week, Anthropic goes to war with the Pentagon. OpenAI closes a $110BN round. Block lays off 40% of the workforce.

   3. JLJason Lemkin0:27

      I think we got the SaaS apocalypse all wrong. Every single CEO I talk to doesn't think they need 40% of their team.

   4. RORory O’Driscoll0:33

      The lesson is never underestimate big markets and momentum. The knife fight doesn't start until the TAM is, like, 60, 70% saturated. The prize for winning is to reinvent the company from scratch and the product from scratch every six to nine months. Congratulations, it's a fun game.

   5. HSHarry Stebbings0:49

      Ready to go? [upbeat music] We are back. Rory, last week this man predicted one of the biggest layoff announcements in history before it came. He is that good at predicting the future. So this week's gonna be even better.

2. 1:16 – 17:01

   Anthropic vs The Pentagon: Who Wins

   1. RORory O’Driscoll1:16

      [laughs]

   2. HSHarry Stebbings1:16

      We're gonna start this week in Anthropic with, uh, [laughs] the Anthropic going to war but being banned, that issue with the Pentagon. Rory, do you want to provide some context for everyone? What's actually happening with the battle between the Pentagon and Anthropic before we dive in?

   3. RORory O’Driscoll1:35

      Okay, yeah. Reminder of context. What happened is that Anthropic actually has a $200 million contract with the Department of War, and in a negotiation, um, ab- about that contract that culminated in a rupture last week, Anthropic was seeking to impose a couple of conditions on the use of their model that the Department of War was not willing to sign up for, and specifically the two issues, as I understand them, and again, I want to preface all this by saying unless you're in the room, you should assume that everyone who was in the room is spinning a version of this that makes them look good. So with that caveat, the two issues appear to be Anthropic wanting to specify two things. One, that you could not use the software, their, their models for, quote-unquote, "mass surveillance," and secondly, you could not use their software for autonomous weapons. Those were two restrictions they wanted to put on the use of their models, and the Pentagon's counter was, "We want to be able to do anything that is," quote, "legal." And that was the bid-ask spread, and it didn't converge, and on Friday afternoon, the Pentagon broke off the discussions and effectively threatened, though as yet I understand it hasn't been done, to i- to, to, to at a minimum cancel the Anthropic deal, but also th- the supply chain threat such that other vendors selling to the Pentagon or even to the US government couldn't use them. So depending on how that threat is interpreted, it's either a mild loss of a, a contract or something approaching thermonuclear, and it's not yet clear has that happened, or if it happened, what it means. So that's kind of the facts and circumst- that's the facts simply put, as distinct from the editorial comment.

   4. HSHarry Stebbings3:26

      Was it the right decision by Dario [laughs] to have these principles and to stand by his beliefs in this way? 'Cause Sam has since swept in and done a deal, and we can get to that repercussions on the consumer side, but Sam has done a deal in replacement.

   5. JLJason Lemkin3:44

      I think he got swept up in a tough position, which is his team, his employees. Because I think Dar- Dario clearly since the day he was lead safety researcher at OpenAI has felt profoundly, uh, important about safety. In fact, they refused to re- release Claude before ChatGPT because he didn't think it was safe. They only released Claude under, under duress. Finally, he goes to his team and says, a while back, and says, "Listen, at least we can make weapons safer. Like we, it's going to happen one way or another. There will be AI in weapons." This is my understanding. And he convinced the team that what we will do is we will make them safer, and that was the justification and, uh, to a lot of folks in AI, it's even Elon said a c- a little while ago, "I, I wish AI hadn't happened, but since it is, I'm all in on next AI," right? So if these AIs are gonna be out there in the military, we'll make the safe one, and I think it just got pushed to the point where I don't... I, I think he has his personal beliefs, but I don't think he'd go to his team and say it was true anymore. It was just pushed too far to say at least we'll make it safer than our competitors were. At least this will be safer than xAI or ChatGPT or some tortured open source thing, because at least we'll-- the good guys will be in there, and we'll make sure we don't do the wrong source of mass surveillance, and that when we do target folks, which we've been doing with some form of AI for years, at least we'll minimize the collateral damage. I don't think he could say it with a, with a, with a straight face, so he said, "I, I, I, I've got to step out on this one. Like this is... I c- I'll lose my te- not only do I not believe in it, but I'll lose my team."

   6. RORory O’Driscoll5:13

      If that's the framing, then I can answer the question concretely. I think you're right, Jason, which means he was wrong. I totally agree. One of the organizing principles of Anthropic has been this, you know, al- [laughs] almost messianic belief in AI, the threat of AI, as yet the willingness to develop AI, 'cause they're the only people who can be trusted with AI, right? I find that weird, but I'd be an idiot not to recognize that this has been an extremely successful unifying principle of this company, right? Which is why they still have the seven founders, when the other guys across the street have lost almost all but one or two of the founders, right? Whatever it is that's in the water at Anthropic, it's working and it has created unity, right? And he would be a fool in terms of running his business to blow that unity, and the man is clearly not a fool, right? So I get it, Jason, you're right. He had this-Belief around AI safety that is not just a lightly held belief in this company, it's a core organizing principle, and arguably even the engine that has driven the enthusiasm to build that company. Given that, I think it was naive to try and sell to government at all, at Department of Defense at all. Because if, 'cause let's, 'cause they too, unlike a lot of us who are just ordinary people, have an organizing principle, which in the case of Department of Defense or Department of War, is defend the United States of America, right? And they believe correctly that that's their task that they've been charged with, and the idea that someone's gonna waltz in and say, "I want a $200 million contract, but I'm going to tell you the safe way to use this," is absurd. And I think the DOD Department was absolutely right to say, "We are not signing up for some bunch of guys in California to tell us what to do. We are the Department of Defense, empowered by the president and the Constitution to defend this country. Go away, little boy." So I think they were entirely right. If you wanted to not sacrifice your principles, you're probably... And if you believe that, you know, that you have really nuanced opinions on how AI should be used in war, then you probably shouldn't sell to Department of Defense because they really don't, correctly, don't give a damn about your principles. And I know, I, I get it, I feel weird saying this 'cause I actually think I love Dario's writing and his speaking. He's some of the clearest, best writing I've seen on almost anything. And I happen to know 'cause I read, as a big believer in Richard Rhodes, The Making of the Atomic Bomb as a book they share, right? And when you read that book, the first conclusion you should come to is General Groves didn't give a rat's ass about the scientists. He humored them, he gave them what they needed, he let them not wear uniforms 'cause it made them happy. But when it come to the crunch, they weren't even in the room when they decide to use the bomb. And thinking that you're not, thinking that you're gonna have the luxury of getting to be part of the decision is unrealistic, naive, and actually constitutionally wrong. I may not love the current administration, but they were elected by 80 million people, and as, and Harry will remember, Tony Benn, the English politician, used to say a cabinet minister, a lib, a very left-wing socialist whom I don't like in any dimension, but he had a great phrase for the civil servants. He turned to them and say, "And who voted for you?" And who voted for you, guys at Anthropic? 'Cause 80 million people voted for us, and the Constitution says we're here to defend the country. So I think drifting into this situation... Now, drifting into this situation, given that that's the core organizing principle, is probably a mistake. It might work from a business and a marketing perspective 'cause you've got a huge bunch of outside lift from it, and on that basis it's been great. But, this is the big but, you're now a little bit at the mercy of how the administration wants to handle this, right? And the thing that's in your favor is they often have a to- short attention span and they move on. And the thing that's against you is they're quite hard-nosed sometimes about grinding you down, and if they want to grind you down, they can. I don't think they should. I think they would be wrong. I think they should say to Anthropic, "You don't wanna do business with us, that's fine. We'll keep our 200 million, but we won't grind you down. We won't declare your supply chain risk. We won't try and destroy your business." But you've drifted into a situation where it's at least plausible they can. I don't think that's a good idea.

   7. HSHarry Stebbings9:30

      If you were an Anthropic shareholder today, would you feel more or less confident today? You mentioned the lift for contacts. They are now number one in the App Store, ahead of, uh, ChatGPT for the first time. They have a consumer brand which actually consumers care about really at mass for the first time, and Dario has presented himself to be a very thoughtful and progressive leader that cares about safety. As a shareholder, would you be more or less happy?

   8. RORory O’Driscoll9:58

      Same expected return, wider variance. By definition, the risk's increased, and therefore you should be mild- you should be unhappy. You were c- I mean, you know, even despite the lift, you were doing really well a week ago. You were clearly the dominant company. You made entire stock indices move at your will, right? You write a blog post and every legal tech company goes down 20%. You had a good thing going, as they say in that meme, right? And now you've drifted into this situation where you've got this ongoing fight with the government, who has way more power, as you're suddenly discovering, than you do. I don't think you're happy about the increased risk in your deal.

   9. JLJason Lemkin10:39

      I, I don't, I think we're entering an era, we've entered an era where no one really cares as long as the numbers are there. No one cares. There's fewer ethical concerns. There's fewer anything concerns. If I just got my March 1 investor update at Anthropic, and we just crossed 16 billion, um, you know, I, I, you know, I, I, I m- I, my email back would probably be, "Great job!" [laughs] With three exclamation points and maybe an emoji or two if it were on my phone, right? I mean, this is the world we live in. So we don't, we don't have the rights, we don't have... I don't think almost none of the money in Anthropic has any rights whatsoever. I would say with most startups it doesn't, no longer matters what I think as an investor. It literally doesn't matter. O- only the struggling ones does it matter what I think. [laughs]

   10. RORory O’Driscoll11:21

       And then you don't wanna care.

   11. JLJason Lemkin11:23

       Yeah. Then, then I have my board partner. [laughs]

   12. RORory O’Driscoll11:26

       But th- you cynic. But no, I, to be fair, I think the question is not do they care about your opinion, the question is how, how would you feel? And I just, I, and I think you're right. If the num- if numbers go up, you're fine. But I think you just have to accept the fact that in return for a significant boost in popularity, good, you have a mild existential risk that probably you can beat in the courts, but nonetheless, if the people on the other side of the table decide they wanna be vindictive, can make things very hard for you.

   13. JLJason Lemkin11:59

       Just one thing that j- I was thinking of to highlight if we get to Block later, right?

   14. RORory O’Driscoll12:03

       We will.

   15. JLJason Lemkin12:03

       Yeah. The, the, if we talk about the, the power of labor versus management, and I know we don't like these terms. They sound very Smithian and stuff-

   16. RORory O’Driscoll12:10

       I like them

   17. JLJason Lemkin12:10

       ... but it's true. At Anthropic w- and, and, and OpenAI and, and we're at the extreme end of power of labor. I mean, people are l- leavingYou know, eight figures of vesting stock at OpenAI because they, they j- they want to work somewhere else after 13 months or 6 months. I mean, this is unprecedented. Even at the hottest startups, people used to stay, okay? On the other hand, for non-AI folks, for the blocks, we have the least power versus management ever. Like, you are, you're less than fungible. You're not even wanted, right? And so there's a subset of folks... And, and listen, Anthropic's probably at 15 billion. This is an important subset, right? There's a subset of folks, it's very weird, where labor has all the power, and, and Anthropic has done the best job of anybody. A lot of the ex-AI founders are gone, too, right? This is like 2021 all over again. You've got to do whatever it takes to retain the team. Like, everything. If they want you to take on the Fox News dude, Pete Hegseth, you do it. If they want you to tap dance on the roof of the company, you do it to keep those, to keep the researchers, right? What Maggie told us, I didn't even know this, when we were in London with Harry, at OpenAI, they don't even let the salespeople talk to the engineers, the researchers. They're not even allowed... Their cards don't work at the building of engineers because they don't want to be bothered by the go-to-market team. They're not even allowed to attend meetings in, [laughs] in the, in the good building. They've got to go to the building next door. [laughs] That's protecting your, your ta- talent at all costs, right? So that's why I think Dario had no choice. I don't think the team would tolerate anything else.

   18. RORory O’Driscoll13:36

       First of all, I love the labor capital distinction, and, you know, you said it, uh, it's, uh, it's really more Marxian than anything else, and I'm gonna go with it. You know, y- you had capital versus labor, and I think you're right. It's super interesting to see in a situation like Anthropic, capital is weak and labor is strong, because those are the most talented people on the planet who have the skill that you need the most. And you're right, in many other organizations it's weak. But I'm gonna remind you, if we're gonna do political economy, right, not politics, but political economy as taught to you here-

   19. JLJason Lemkin14:04

       No, I'm gonna defer to you here. Yes

   20. RORory O’Driscoll14:05

       ... there is a third actor. It's called the state, and I think what you're seeing very clearly, and I just want to say it very clear, is the state is more powerful than Anthropic, and I think that's the lesson you gotta learn. You can have, I think, you... Yeah, and it's interesting. If you listen to some of the... One thing that some of the folks talking about AI even two or three years ago were hinting at this, where they're like, "Will the state intervene?" And I think the state has extraordinary powers because it's the state. That state has, as Weber points out, the monopoly on violence within a geographical area, right? A- and I think that i- in terms of appeasing your employee base, you gotta do what you gotta do, but drifting into getting into a conflict with the Department of War on something where I think logically you're in the wrong because you're trying to interpose yourself over and above the people who are charged with doing this is, is just a bad place to be, and I would want to tiptoe back out of there. And if it all it costs you is a $200 million contract in the context of a $14 billion business, you should declare a win and move on. Because if the state deci- I mean, for all this talk about AI being so scary, let me tell you what we learned in the last week. AI might be scary in the theoretical after, "Oh, it could be used for this, it could be used for that." The state is scary in the real sense of, we have laws, we have men with guns to enforce those laws, and we can take your company if we want to. And I bet you, even though you look at those laws like the Defense Production Act, you look at those laws like being declared, what was it, a supply chain risk, and you kind of go, "I think I could win that one in court," it's still a sobering moment when you realize, "Hmm, the person in charge of the most powerful military on the planet thinks that you are a supply chain risk, wants to invoke powers granted to him by Congress to do that, and you've picked on them. Hmm." I would have preferred to skip that.

   21. JLJason Lemkin15:57

       And, and you've got David and the rest coming straight for you nonstop, right? It doesn't help.

   22. HSHarry Stebbings16:02

       When you're talking about the power that labor holds, what you're suggesting with that, given Sam sweeping in and taking it, is that he does not face the same pressure from his employee base-

   23. JLJason Lemkin16:14

       No, he did. His team hated that he did it, right?

   24. HSHarry Stebbings16:16

       His team hated it.

   25. JLJason Lemkin16:16

       He had to immediately say they're gonna, that he's going to unilaterally change the terms of the contract the next day. I, I mean, he's a very smart guy. Uh, he's signed a few contracts in his day. That was numb nuts thing to say. I, I mean, I, I know he believes it, but, "Oh, we're just gonna change the clauses ourselves, uh, on ChatGPT so that they say what we want them to say." I mean, good God.

   26. RORory O’Driscoll16:38

       I, I agree with Jason there, Harry. I think what the last four days have shown is the power of labor at OpenAI, too, and it's ironic because the first time we saw the power of labor at OpenAI was the power of labor to insist that Sam comes back, which was, you know, back in 2023 when you had that magic moment when everyone... When, when labor basically said, "We want to pick our boss," and the owner says, "You're right. You should

3. 17:01 – 23:29

   Was Sam Altman Wrong to Take the Deal

   1. RORory O’Driscoll17:01

      pick a boss." Right?

   2. HSHarry Stebbings17:02

      So was Sam, so was Sam wrong to then do that deal in your mind? You said it was naive of Dario to maybe enter it. Do you think Sam should have not entered it also?

   3. RORory O’Driscoll17:12

      It's the most entertaining of all possible outcomes. Um, he's probably right on the merits, and as often happens in life with a certain kind of person, it's only when you're absolutely right on the merits. It's the one thing that, I mean, when you actually believe what you're saying, and I read what he's saying about y- in a democracy, you know, we as the, the final arbiter should be the government, not individual people. I actually think he was right on the merits on that, which of course means pleasingly it's the one thing that's going to bite him on the ass, right? My, you know, i- if you see what I'm saying, is that ironically, the one time I think he was, like, entirely correct in terms of how companies should interact with the Department of Defense, it's actually going to be the... And I think actually signing the contract on the merits wasn't awful, but you're right. His labor doesn't want him to do that. So I think in terms of opening Pandora's box in his company, he was wrong.

   4. JLJason Lemkin18:06

      It's to- I think he saw, I mean, he wasn't even approved at the Department of Defense for this application, right? Only xAI was, and they had lost. They had lost because Palantir picked Anthropic. They'd already lost last summer. They'd already lost this, this, this war, and he saw all of a sudden there was a break, which happens to every founder, right? They u- There are these moments in time when your competitor goes down for a week or screws up a key deal-And he did what most of us in B2B would do. He pounced on the deal. He immediately shot the email to the CEO. You know, "Rory Co. was down all weekend. I can have you up on our infrastructure on Monday." [laughs]

   5. RORory O’Driscoll18:40

      I'm actually gonna argue something that's gonna sound naive. I actually think it wasn't just a pounce while the other guy is weak. It actually was a roughly trying to do the right thing. And actually, as I say, I think the odd thing is I think he was right on the merits, right? It's like, we should be willing to sign deals like this because the government should decide, but his labor's not gonna consent to that, right?

   6. JLJason Lemkin19:00

      I think it's just everything is accelerating in the capabilities of these models. Everything has accelerated since December, right? Everything's gotten better. So it's just very hard to predict for a military application what these models will be able to do in 12 months. It, and it could, it could literally could be, uh, beyond what we could conceive these LLMs could do. And I'm not saying that that is a, a bad thing because there's bad actors in the world, but even understanding where it's all gonna go, it's hard to predict. Hard to predict. Yeah.

   7. RORory O’Driscoll19:29

      Just, just to ask a political economy question-

   8. JLJason Lemkin19:31

      Yeah

   9. RORory O’Driscoll19:31

      ... who should make those decisions on how to use this technology? A privately held company or someone elected pursuant to the Constitution of the United States of America with 80 million votes and a Congress? Who should make those decisions if they are quote, unquote, "on the same page"?

   10. JLJason Lemkin19:43

       They're all-- Well, it's tough. I mean, I don't want to spend all our time. It's tough because there's no good answer.

   11. RORory O’Driscoll19:47

       Anyways-

   12. JLJason Lemkin19:47

       We know the last thing we want is Ro Khanna or whoever he, the version is on the right, making a hyper-political decision. We, we don't want that. We don't want it politicized, right?

   13. RORory O’Driscoll19:57

       J- J- No, I disagree. We do want political-- How do you think we make any fucking decisions? It's called a Constitution for a reason. We elect people to decide for us.

   14. JLJason Lemkin20:02

       I don't think we can-- I don't think most of our political leaders even understand how AI works. I don't, I don't think most of, most of the folks in general, even in on X, understand how it works. They don't, they don't understand.

   15. RORory O’Driscoll20:12

       Hard-nosed comment again. You might think probably-

   16. HSHarry Stebbings20:14

       Well, look, guys, there's lo- there's logic and there's rationality. You're logically right, Rory, but Jason is rationally correct. The majority of-

   17. RORory O’Driscoll20:21

       No, I disagree.

   18. HSHarry Stebbings20:23

       The majority of governments around the world are not sufficiently versed in artificial intelligence and deployment and evolution of it to make sound decisions. I think we would all agree on that.

   19. RORory O’Driscoll20:33

       I wouldn't. Look, by definition, if you're running the government, on any individual thing, experts know more, right? Your job is not to be the expert. Your job is to be able to manage the experts, right? And I think it's the same mistake, frankly, that folks made in COVID where they did, they defer to science. No, it's listen to science and make informed judgments. Now, you might argue that the people making those judgments at various times have been mediocre or even bad at judgments. But the way to solve that is to elect different people to make judgments. You can say, "I don't trust the US government, so I don't want to deal with them." That's a totally rational thing to do up until the moment when they invoke the Defense Production Act, right? Totally rational. But what you cannot do is say, "I'd like my $200 million, but I'd like to tell you what to do as well."

   20. JLJason Lemkin21:18

       Just one interesting thing on the deal, just structurally, putting aside and then, and then we can move on. I don't know exact- Like, it wasn't being negotiated before it was being signed, right? By, by the nature of any big deal. So it was signed last summer. That means Anthropic was probably closer to a billion-dollar run rate and still trying to break it, 'cause it was a billion dollars in January of, of 2025. So signing a $200 million deal plus 200 million from Palantir pull through, okay, plus blocking your competitor from Palantir and others, that was a probably a pretty big deal at the time. That was probably one where, like, he was honest that he swallowed some of his principles, like raising money from sovereign wealth countries. But it was so important to winning in the enterprise and building this cloud, it probably actually made more sense to take the risk at a billion-ish, right, than at 15 billion, where you're like, "Oh, you know, everyone's gonna be like you throw away the 200, 400, 400, 400 with Palantir," okay? 200, 200. Less of a, a har- tougher decision at a billion run rate. Like, most of our portfolio companies, if they came in and said, "We can grow an extra 40% this year if we sign a contract we'd kinda rather not sign, but it's totally legal. And instead, Rory, instead of growing 60% this year and having no exit, we can grow 110%," I think most investors [laughs] would kind of say, "Take, take the deal. Swallow your pride," uh, right? And so it was a bigger deal at the time than it is today, I think, right?

   21. RORory O’Driscoll22:39

       Agreed. But I actually genuinely, now having been harsh, I'm gonna compliment. I don't think they did it-- I don't think this happened because they needed the money then and they don't need the money now. I think, like often happens in life, a bunch of peop- uh, people drifted into a situation that nobody fully thought through, and then, you know, uh, y- y- you end up at this point, even though you didn't plan to be there. I'm su- I think actually the relationship happened initially via Anthropic, sorry, via Palantir. So, uh, they were selling to Palantir as one would. A great customer to sell to. Palantir sells to the government. You get dragged in, and you probably sit there thinking, "Well, they'll be reasonable. We'll go and talk to them about how we operate." I mean, the Wall Street Journal covered it today. Part of this, as is always the case, is personality clashes. You're like, "Oh, this is the way I look at the world, and this guy at the other side of the table looks at the world totally

4. 23:29 – 27:44

   OpenAI's $110BN Mega Round: The Breakdown

   1. RORory O’Driscoll23:29

      different."

   2. HSHarry Stebbings23:29

      Final thing on, on OpenAI, Anthropic, and then we will move off them. The ultimate deal maker that is Sam Altman closes $110BN round, four X the size of the largest IPO ever. I mean, Jesus, you've got to give the man credit. I mean, what an absolute machine.

   3. RORory O’Driscoll23:48

      Totally.

   4. HSHarry Stebbings23:50

      Anything other than plaudits? Is there any money left in the world to fund these businesses? The 200 and 400 million in revenue seems relatively paltry compared to the $110 billion he just raised.

   5. JLJason Lemkin24:02

      Yeah, I mean, th- there's some meta que- I mean, obviously, they're, they're not public, and obviously, the benefits are inuring to a relatively modest amount of shareholders, right? All of that stuff. But it does make you wonder if we should even be talking about the IPO window or any of this crap. It doesn't even matter when OpenAI can raise 110 billion. Oh, we're all worried some B2B company can't IPO at 400 million, growing 40% at 3 billion. What difference does it [laughs] make? It's so minuscule. I mean, it matters to the humans that work there, but it's not even relevant to the economy when OpenAI does 110. And they're gonna do another one.Before the IPO, probably before the IPO if they can, right? May- maybe not. It'll either be an IPO or another round like this, but it's, it's, it's a f- it's a force of nature, right?

   6. RORory O’Driscoll24:43

      It's clearly a force of na- and it is astonishing. I mean, you know, any way you cut it, it's like, you know, you add up venture dollars raised year to d- i- invested year to date, you know, you kinda got 30 on Anthropic, 110 if you call it VC. It's 140. You know, it's more than US venture for all of last year. So it, it, these are astonishing numbers, right? I do wonder, and you're right, i- i- it's a very odd circumstance to have a private round be 4X the size of the largest IPO ever, right? And it does make you wonder how that IPO gets done, right? And we've talked about this before, right? And it's interesting, there's little nuances here. Amazon's, I think, 40 or 50 billion, some of it's up front, some of it's commensurate on either an IPO or AGI, which is just one of those weird things, that some of the money only has to go in once the company either achieves AGI or goes public, right? Which, to me, and Jason, you mentioned a while back, exudes a little bit of the IPO support coming in. In other words, is this really 15 million now and 35 million as kind of a placeholder for the book on the IPO or something?

   7. HSHarry Stebbings25:50

      There are also precedents of, you know, NVIDIA investing 100 billion, and then the kind of walk back of up to 100 billion. I- is it that again? I'm, I'm genuinely confused.

   8. RORory O’Driscoll25:58

      No, it's not quite that, because NVIDIA talked about investing 100 billion. You remember I, I typed out the pr- I printed out the press release, and then they ultimately invested 30 billion. So they, you know, folded back slightly. Amazon came in for 50, but of that 50, it's not like it's a press release. It's actually a real commitment, but it's commitment on closing conditions that are either an IPO or AGI. So they're, it's not money now, right? If you needed to pay your workers next week with that 35 million, uh, billion, it wouldn't be there.

   9. HSHarry Stebbings26:28

      Wh- when does this go public?

   10. RORory O’Driscoll26:30

       You've gotta believe that the next round for this, Anthropic, and SpaceX are all public offerings. Articulate who writes the next check privately and what the investment thesis is. I don't think there's more capital there. I could be wrong, but it's hard to imagine another round after this. So you just therefore, by a process of elimination, at some point you gotta go public, and the next round probably is a public round.

   11. JLJason Lemkin26:52

       An important point on an asterisk that we have a double asterisk to. So it's 110 billion round, but 50 billion from Amazon, of which only 15 billion's up front, right? The rest is AGI or IPO. But Amazon's own free cash flow has fallen to 11 billion a year, 'cause it's spending so much. So we are literally- OpenAI had to go to Amazon, and they had to go to everybody on planet, NVIDIA and Sa- and, and they're exhausting all... Like, Amazon doesn't even real- does not-- It has the resources, but it does not have the free cash flow to fund its commit, right? So it's not that, uh, could Amazon do 150 billion, 300 bill- I mean, we're re- there's a limit here where NV- where the folks that did this round literally cannot do any more, right? And this is probably why Jensen walked the 100 billion back was there's a theoretical idea where it makes sense, but the world has changed, and it's, it's, that's, that's a lot of free cash flow, even for NVIDIA, 100 billion, right? So we're reaching their limits.

5. 27:44 – 34:42

   Who Has a Bigger Valuation Premium: Sam Altman or Elon Musk

   1. HSHarry Stebbings27:44

      I know this is an absolutely stupid, probably, suggestion. You can both fire me for it. Is there a Sam Altman premium? We've spoken about the Elon premium before. [laughs] If Sam were to say, "I don't wanna do this anymore," is there a Sam Altman premium?

   2. RORory O’Driscoll28:00

      I don't think in the same way. His genius has been deal-doing and the oomph and the chutzpah to raise 200 billion-plus privately held. The, you know, and he's obviously technical. He has a computer science degree f- from memory. I wouldn't swear to that, but, but he's, but Elon's genius at the margin has been the ability to pull off amazingly complex engineering pro- projects sequentially. So I don't think even, I don't think there's the same premium there, and you could even, t- t- to make, to, to make it a little more pointed as we're against Elon just for a second, OpenAI trades at 700 billion pre this round, but that's 'cause it's hyper-growing at, you know, I can't remember the grow- from 12 billion last year, from 3 or 4 billion in gap revenue to 12 billion, 20 billion ARR run rate. It's, you know, 30 or 40 times revenues for hypergrowth. Astonishingly, I continue to point out, Tesla continues to trade at 10, 12, 13 times revenue for declining revenue, right? The Elon premium is way higher [laughs] , you know, relative to the performance of the underlying asset. You ask who's got the better premium, Elon or Sam? And actually, the easiest way to ask that question is to say to yourself, remove the person with the premium, right? Tesla trades at a trillion today. I think if Elon died tomorrow, they'd trade at 200 billion. OpenAI trades at 800 billion today. I think if Sam Altman died tomorrow morning, we'd trade at 600 billion. They'd go, "We better get someone else to run the thing." Brett Taylor would be in charge, right? And Brett Taylor would be great, and he'd figure it all out, and they'd build a company and go public. If Elon goes, who's gonna make the robots, and who's gonna make the robotaxi? And if you don't make one of those two things, you've got a car company with a declining car product line, flat revenue growth in a tough market for electronic vehicles. You'd be down 800 billion bucks.

   3. HSHarry Stebbings29:48

      I think that's fair.

   4. JLJason Lemkin29:48

      Yeah, you-

   5. HSHarry Stebbings29:49

      I do

   6. JLJason Lemkin29:49

      ... if Sam left, you buy b- you buy, um, Sierra for 80 billion, 10% of the value-

   7. RORory O’Driscoll29:54

      And you let him in

   8. JLJason Lemkin29:55

      ... um, you buy out the sharehol- you know, whatever the math works, whoever gets what. You do, you just do it overnight for 80 billion, and your problem is solved, and, and possibly the company's stronger afterwards other than the fundraising. It might be better.

   9. RORory O’Driscoll30:05

      Yeah, agreed.

   10. JLJason Lemkin30:06

       And if you have one of the greatest technical CEOs of, of our lifetimes in charge of it, um, instead of a very brilliant but ultimately non-technical founder, I mean, it's, it's tough to beat, it's tough to beat Brett Taylor's background. It's, it's, and he's already on the board, right?

   11. RORory O’Driscoll30:19

       But you know what? Fundamentally, Jason's right. You are right. Fundamentally, you'd be like, "Ah, we're just gonna build a company now. We're gonna build ChatGPT. We're not gonna have distractions. We're just gonna be fine." You're exactly right.

   12. HSHarry Stebbings30:30

       I'm gonna push you both before we move on.

   13. RORory O’Driscoll30:32

       Sure.

   14. HSHarry Stebbings30:32

       When does it go out, and what price does it go out at?

   15. RORory O’Driscoll30:35

       I think there is now a subtle, unspoken rush to the money that is taking place between those three mega cap companies. Everyone else is kind of to a rounding out on the sideline until these deals get done. I personally would say the sooner the better for everyone, right? So I'd love them to go public as quick as can and get it over with. The question is not when are they go- wanna go public, 'cause I think if they could wi- if they could all wave a magic wand, by the end of this year I think that all wo- all three of those companies would wave the wand and say, "Let's be public. It'll be, it'll be safer."

   16. HSHarry Stebbings31:01

       I'm gonna say October, and I'm gonna say it's gonna go out at $1.5 trillion.

   17. RORory O’Driscoll31:06

       But OpenAI.

   18. HSHarry Stebbings31:07

       Mm.

   19. RORory O’Driscoll31:09

       I think you can trade that in polyma- on, on some of these kind of tokenized bets. I mean, you know, which probably sh- should be something we should do.

   20. JLJason Lemkin31:16

       It's a logical bet. Um, they might, they might actually seek initially a lower price, um, to not stretch it for an IPO and then see what happens. Uh, the, the SpaceX IPO price that Elon is making up, $1.75 trillion, I mean, it's a crazy number, but it's not 2X the last round, right? So there may be... It, it may make sense to have a modest step-up, but that's really just aesthetics. You know, it, the market will settle. It would, um... Uh, I, I do th- I do think thinking on it, and listen, I could be wrong on this one, but thinking on this live now, I do think there's risk that they've exhausted some pools of capital from, from, from the, the, the round-trippers, the SoftBanks, the NVIDIAs, and now Amazon. Like, a lot of that money, th- the fact that so much of the mo- money of Amazon isn't in the bank today, to me it, it doesn't count. Like, maybe it sets the IPO up, to Rory's point. Like, that's great to set up to have $35 billion of your IPO pre-sold. Like, that does make the IPO easier. Like, let's, let's be clear. Like, like, that, it's a gift walking in and have half your book sold. But if they've exhausted the money, then yeah, they should go public in October. I, I think it's right.

   21. RORory O’Driscoll32:18

       Just to comment on the valuations. I mean, I, I, I'm not in the, "Oh, it'll be fine at $1.5 T..." I, I think those are such huge numbers that they're very much predicated on the overall market continuing extremely strong. I mean, you know, S&P, despite all the noise within the system and all the SaaS apocalypse, S&P is plus or minus 2% from an all-time high and trading at an all, you know, a 20-year high in terms of any kind of trailing KPE basis. So this is selling securities at a time when people are hot to trot and buy securities. So getting it done in today's market, who the hell knows how it gets done? It could easily get done at a, an incredible price. But it is worth pausing and just looking at the vast distance between the valuation on any kind of fundamentals basis and the amount of leaning into the future you have to do to get to those kind of valuations. And it's kind of slightly different between them, 'cause OpenAI is a, it's growing fast, really fast. Provided it continues to grow really fast for two or three more years, it all works. So you can pay, you know what, 40, 50 times revenues. I mean, the fact that I even said that sentence [laughs] at scale i- is how that one happens. And if there's an appetite to pay 40 or 50 times revenues at that point in time, you can do it. SpaceX is even harder because you're basically having to underwrite a fair slug of next generation technical risk. Admittedly, you're underwriting with the, with the person who in the last 20 years has proven most able to deliver that technical risk. But you've got the whole Starship risk, and then you've got the next generation, um, Starlink 2.0 direct-to-cellular risk, and then you got the data centers and space risk, and you need all that TAM 'cause the existing business is $18 billion growing 20%. So it's not obvious that it's worth 100 times trailing revenues for 20 per... If that was a SaaS company doing $18 billion, growing at 20% with m- modest profitability, Jason would be sneering and saying, "Five times."

   22. JLJason Lemkin34:14

       I wouldn't be sneering. I'd be saying, "Five times."

   23. RORory O’Driscoll34:17

       Yeah, you would be. Yeah.

   24. JLJason Lemkin34:18

       Yeah.

   25. RORory O’Driscoll34:18

       And the difference between the last-

   26. JLJason Lemkin34:19

       I'd be lamenting it. Like, I'd have my head in my hands, but I would not be sneering.

   27. HSHarry Stebbings34:22

       [laughs]

   28. JLJason Lemkin34:22

       I'd be commiserating.

   29. RORory O’Driscoll34:24

       You'd be-

   30. JLJason Lemkin34:24

       I'd be like-
6. 34:42 – 43:09

   Why We Got the SaaS Apocalypse Wrong?

   1. HSHarry Stebbings34:42

      You said if it was a public company doing, you know, 18 at 20%. Jason, before we started recording you said something, and I wrote it down. You said you thought every public company would miss their numbers.

   2. JLJason Lemkin34:53

      [laughs]

   3. HSHarry Stebbings34:54

      Can you just exp-

   4. JLJason Lemkin34:54

      I think they're just, I think all the publics... I mean, I don't mean to be negative. I think all, almost all the public software companies are just going to do worse and worse for the rest of this year. I think, you know, Mon- Mongo was down, what, 20... The largest drop ever after a strong quarter because they, they lo- they, they, they had s- they said they're gonna, gr- growth will drop back to the teams, right? Even though they crushed their quarter. Um, you know, you had Iran from Monday on the other 20VC this week, who was great, but the, the public markets think that they're gonna keep, uh, uh, they're gonna keep re-rating growth lower and lower each quarter. I just don't see any positives for the existing group of B2Bs. I, I... And, and Rory's point is we just need more good ones in to raise up the median and the numbers, and, and maybe it's that simple. But I, I actually think, I think we got the SaaSpocalypse all wrong. I, I, I, I, I think that there's, it's, um, it's not vibe coding that, that's killing us. It's just we, we, everybody has lost the, the, the, the, the way to growth. And I think the, so, and I think because of NRR, because of revenue retention, because so many of these numbers are backwards-looking, I think e- almost every one is worse than it looks. You know, it's funny, I wrote it up this week. You could expand the basket very widely, but for the stocks that I follow, the most successful one is DigitalOcean. DigitalOcean's up 28% this year. DigitalOcean had all these years to, uh, to a billion in cloud revenue. Like, in some ways very impressive to a billion, right? But on the other hand, like, w- you know, it's, it's pretty good business to only be doing a billion, right? You're pretty... You're, it, it, you could be critical, but you need to be radically accelerating growth and profitability at the same time, and you're looking at these public eyes and you're like, "Who the hell is gonna radically re-accelerate growth?"Who, who the hell? And that's why when I have to come up with my four to bet on, I could only find three. Like, I'm still gonna do four, but I could only find three because I can make the, the p- the case that this one's undervalued and that the markets are, are... don't see the, the inherent value in, in, in... But the markets are saying, "You guys got to re-accelerate," and they're all gonna de-accelerate. And so I just think the p- almost all of these public B2B companies are in worse shape than they look. And while I wish we didn't have the SaaSpocalypse, man, I just, I think it's worse than vibe coding. I think vibe coding is one of the most minor threats to software companies in our lifetimes, and I vibe code constantly. It's a minor threat. And that's why they all have to cut half their teams, to our conversation last week. That's... I sounded so pressing. What are you gonna do when, for the next two quarters you keep... Like, you may even make the quarter, but you keep having to drop your earnings estimate and rate down. At some point over the course of the year, you're just gonna have to cut half your team because otherwise you can't make the math work.

   5. RORory O’Driscoll37:36

      I wanna talk about the team, but before I do, I just wanna put one caveat out there, right? You do have to take... I mean, you do have to take price into account, right? And well, let me tell you what I mean by that. Yamango had a very strong quarter, modest delta in terms of future guidance versus expectations, dropped 25%. Why? Answer: It was trading at 40 times EBIT, next year's EBITDA. When you're trading at a super high price, uh, eight or nine times, um, GAAP revenue, right? When you're trading at a super high price and you go from 30% growth to even, you know, I think it was the guidance was 23, 24, clearly assumption would be 24, 25 would be actual result. When, when you're trading at 40 times EBITDA, it doesn't take a lot to knock you off your pedestal, right? Conversely, a whole bunch of these things are now trading at seven, eight to nine times next year's EBITDA, where even 8, 10, 12% growth can probably make it a, at, at the very least, highly unlikely to decline by 30%, and arguably more upside than downside, right? So I do think-

   6. JLJason Lemkin38:39

      But why will they re-accel- I just don't believe these-

   7. RORory O’Driscoll38:41

      I didn't say they re-accelerate, Jason. I think there's two separate things.

   8. JLJason Lemkin38:44

      Yeah.

   9. RORory O’Driscoll38:44

      And it gets to your Block comment. You've got to do one of two things. I do agree you're right about what... You gotta do one of two things. If you re-accelerate enough, then they'll cut you slack on EBITDA, on, on profitability, provided you have some either profits or convergence on profits. If you don't re-accelerate, then you're right, all you're selling is a profitability story, and then we're gonna segue straight to Block. The only way to do that is to look at your cost structure and say, "If I ca- if, if my..." I mean, I hate the rule of 40, but let's put it out there. If my revenue growth is 10%, I can subtract 10 from 40, and I come up with 30% free cash flow required. So you're right, you start running the math.

   10. JLJason Lemkin39:24

       It's just brutal because there's been... The amount of spend that has gone to cloud and LLMs is like nothing we've ever seen before, and if you're a public software company and you've captured none of that, it's a disaster. Like, it's, it's a Titanic. It's worse than it looks because you can't figure out any way to take all of that spend and turn it into an agent and LLM that your, your 10,000, 50,000, 100,000 customers wanna pay you more for. I mean, it is the biggest fail and own goal in our history, and it's gonna get worse. It's go- There's so many public folks who are, "We're rolling out a beta," or, "We've added a cloud connector this week," and it's a disaster at this point in 2026. It is. And you can see it when a lot of these public company CEOs speak. They don't have the con... Like, I was on the earnings call with Marc Benioff. He's got the confidence. He believes that there will be a lift from Agentforce. He, growth already... Now, some of it is inorganic. Growth is up, and he believes he has a performing agent. Uh, he... Look, I was there in the building. The dude believes, okay?

   11. RORory O’Driscoll40:22

       Yeah. No, I hear you.

   12. JLJason Lemkin40:23

       But I can smell the lack of belief in other public... I know Harry can too when he interviews them. You can just smell it out of the pores. And a- as a human being, I have empathy, but good God. That's why I think it's all wor- because of retention, it's all worse than it looks. It's all worse than it looks because you can hide under y- uh, multi-year contracts and annual contracts and price increases and, but the f- like, you, it's too... You had so much ti- Like, these three kids from Stanford that figured out how to use, uh, Opus, why couldn't you? It's, it's out there. It's not even that expensive. Why can't you figure out how to use Opus? Why can't you put your 20 best engineers and just freaking clone Harvey or ElevenLabs or Lovable? How hard is it to clone those apps, guys? You all deserve to be fired. [laughs]

   13. RORory O’Driscoll41:09

       Hang on. I, I, uh, actually, I thought Eoghan at Intercom had a really great piece just this week on, you know, what it took to get the f-

   14. JLJason Lemkin41:19

       Yeah

   15. RORory O’Driscoll41:19

       ... to take his existing business, you know, gut it, and build Fin on top of it. And, and n- and that, that's what I meant when I said, "Oh, uh, Jason, it's hard." Not from a technical perspective. That's part of it, but just from a fortitude to make all those changes, to put all your bet on the new thing. There was probably a year of feeling ridiculous 'cause you're talking about the small thing, and the big thing is so much. Customers are saying, "Why aren't you focused on the big thing?"

   16. JLJason Lemkin41:42

       Yeah.

   17. RORory O’Driscoll41:42

       And he just described very well what it took to be bloody-minded enough to make that happen. And you're right, Jason.

   18. JLJason Lemkin41:48

       Yeah.

   19. RORory O’Driscoll41:48

       I do think you are correct. That's the kind of attitude that any... I mean, the opportunity is there, and it's, I'm interested to hear your take on Agentforce either this week or next week or whenever, but that's the kind of bloody-mindedness it's gonna take to push through. I think more of an innovative thing-

   20. JLJason Lemkin42:03

       You do need it, and I just don't see it. I just don't see it across startups, and I don't see it in the public companies. I don't see... And Eoghan's been... And now listen, now Eoghan had to come back as CEO. There's complexity there too, right? But he's been brutal, that this is a founder market and it's hard. Um, and yes, it was harder for them because they had an institutional base he had to partially retain but also partially abandon. A lot of complexity, but woe is me. Like, uh, his points are all, were all really good, but you know what? Maybe, maybe he got there a quarter or two earlier than others because it was in CX and support, okay? So he could see the change earlier than every- It wasn't just going all in. It was that also he was in one of the two categories that changed the most quickly, right? Um, so, uh, but what about the rest of the world? You've had timeYou, you, you can go out there and say on Twitter, "LLMs are fungible and that systems of record should benefit because, uh, because, uh, we own what matters and, and LLMs are commodity." I mean, give me a break. You've-- Where's your hun- Show me 100 million. Show me 200 million. Show me 500 million. You've had, you've had s- 16 months now.

7. 43:09 – 49:03

   Why Salesforce Could be the Best Buy in Public Markets

   1. RORory O’Driscoll43:09

      I, I, but I think Toby-

   2. JLJason Lemkin43:10

      You've had 16 months

   3. RORory O’Driscoll43:11

      ... I, I think like Toby at Shopify is one of the most brilliant CEOs and technical minds. Uh, I think Gustav at Spotify is too. I'm genuinely naive here. Are you telling me they've missed a beat as CEOs and brilliant-

   4. JLJason Lemkin43:26

      No, Toby is dragging his company mercilessly into the agentic era, era, rolling out their Agentic Commerce. I don't know how successful it will be, but it'll, it'll be front and center of Shopify, right? It's becoming front of center. So, but he feels, to me, roughly at the pace of Salesforce. He probably should be faster, right, because it's a more agile organization. But they're, they're dragging their companies with real product into the future, and I think everyone else, they're gonna-- they're, they're already starting to fall into the, the event horizon, into a state of terminal decline, that their teams don't wanna do the work and that they've-- they're losing-- They're- they don't have the ideas. They don't have the ideas. "Oh, Jason, I rolled out AI SDRs." Great, but you're a software company. Where's your agent? Where's your agent that goes out and does all the work for your customer base, right? Um, they're just headed in the event horizon, and they're trying to put on the jets, but it's, it's, it's getting tougher each day to recover. Yeah.

   5. RORory O’Driscoll44:24

      Well, Rory, we're gonna do a Christmas party, the three of us, and, and Jason's our, our bringer upper. Yeah? He's the mood piece. Yeah, he's, he's gonna be... Yeah, he's gonna be the fluffer upper.

   6. JLJason Lemkin44:33

      I, I-

   7. RORory O’Driscoll44:33

      Yeah, no

   8. JLJason Lemkin44:33

      ... hey, listen, if you're really in the game, no, seriously, if you are in the game, I will be your biggest interrupter. I will use your product. I will support your product. I will say it's great. That's why, that... Honestly, t-t, for a long time, to no benefit of doubt, that's why I was so supportive of Agentforce, 'cause we started using it and it works. So I'm like, "I'm gonna take this example," and it is not perfect. Like, I can tell you all about it, more than almost a- I don't think there's that many people that use 28... Five different, 10 different GTM agents, including Agentforce. We might be N of one, and I can tell you the flaws and the issues and the challenges, but at least they're in the game. Armies of FDEs, armies of folks deploying custom agents, and it will get better. And, and I t- will tell you, it is still changing. Agentforce is constantly changing. It is not static. But like a real AI company, it gets better every month or two. It keeps getting better and better and better like our AI tools are, rather than me sticking to beta. So the, I'm not saying, uh, Salesforce is gonna turn into 50% growth next year, but I wanna see this or better from any- anyone. Anyone in software, you gotta be doing as good or better than Salesforce, or I think you're hopeless. You're hopeless. "Oh, our industry doesn't really like AI. Oh, you know, our cu-" Well, good luck to you. Good luck, good luck to you. I will bring the holiday cheer. [laughs]

   9. RORory O’Driscoll45:46

      I mean, I just wanna reiterate my position in the boring middle here, which is I think the straw man of these public company CEOs are ignoring it totally. It, it isn't really true or fair. I think that... Yeah, I do believe that most of these SaaS companies, I, I don't think they evaporate or go away. I think it's... I, I, I think provided they get with the program, depending on the industry and market, you know. There's the markets where you've got plenty of time where AI is not gonna be widely disruptive from a top-line perspective. I'll talk about OpEx next. All the way to where it's obviously disruptive. I think it's well within the capabilities of highly compensated executives who have their ass on the line to figure this out. Most of the companies that already have scale should not be vanished from this, which is a different statement than saying they will ever regain their, you know, 20, 30X, um, EBITDA multiples, still less, God forgive us, 20X revenue multiples, which they got in 2021.

   10. JLJason Lemkin46:41

       Here's the thing that I don't understand about the SaaSpocalypse, but this is what I think makes these companies terminal, okay? What happened through December 2025 is basically the public markets said, "Listen, we're actually gonna give you a growth premium for ever lower growth." Like the top tier in 2025 was if you grew 30% or higher, you were, uh, uh, I think on average, uh, valued at 25 times revenues, okay? And this held even each year through 2024, our growth slowed, and the markets were like, for whatever reason, the markets would say, "Listen, it's cool. As long as... We will, we will ratchet down what we expect from public software companies. We'll ratchet." And then at some point in January, the market said, "No longer. No longer will we ascribe high growth premiums." And it came out of a, a shot in the blue, and this is why everyone is still stunned now because it wor- this w- like, this guiding, this, this deceleration but multiples staying strong, if you were in the top quartile, it worked for three years, 20 t- 2022, 2023, 2024, and 2025, right?

   11. RORory O’Driscoll47:43

       And we are the facts, not the stunned, 'cause you are zooming out 20 years. But-

   12. JLJason Lemkin47:46

       I, I see them stunned. I think Harry th- sees them stunned too

   13. RORory O’Driscoll47:49

       Yeah, maybe, yeah, but then they just didn't look at the chart, right? If you look at the chart, it's really simple.

   14. JLJason Lemkin47:53

       No, it's your stock price, Jason.

   15. RORory O’Driscoll47:55

       Hang on for a job.

   16. JLJason Lemkin47:55

       [laughs] It's just, it's just, it's just-

   17. RORory O’Driscoll47:56

       No, hang on. Stop.

   18. JLJason Lemkin47:57

       Yeah.

   19. RORory O’Driscoll47:57

       But for, for what... I agree with you on the conclusion, I, where you were at, right? Is there was 15 or 16 years of the bucket growing at an average of 30%, valued at an average of six times revenues. Then there was two years of COVID, bucket growing 40%, average of 20 times revenues. And then you're exactly right. What happened is we de-accelerated, not just back to 30%, but down to 15, and then the market said something stupid like, "Well, we're back to six times revenues. It's okay," like it used to be. And you're right, Jason, it was wrong. It was six ti- once upon a time, you were six times revenues and growing at 30%. Now you're at six times revenues growing at 15, and you sat there for two years where it was like when you do a chart that just shows revenue multiples over time, it's a stupid chart. Because it looked like it had gone back to normal, but you're right, Jason, the, the growth had gone out the bottom. And then all that happened last year is people's eyes were opened, and they were likeIt's not a temporary growth decline because of X, Y, or Z. It's now permanent because of AI. Oh my God, what are we thinking? You should be at four times. That's the

8. 49:03 – 1:02:45

   Block Lays Off 40% of Team: AI or Overhiring

   1. RORory O’Driscoll49:03

      movie.

   2. HSHarry Stebbings49:03

      It's now permanent because of AI. That's a really interesting statement. We obviously saw following Jason's statement, I, I don't know if Jack was listening. I'd like to think that he is and took Jason's, you know, guidance as instructions, and announced obviously laying off 40%. It's kind of all under the hood of AI efficiency improvement. Is it truly that, or is it a very bloated organization that needed resizing? Jack's continuously overstaffed his companies, and it was just a resizing with a mask of AI.

   3. RORory O’Driscoll49:38

      Jason, you should go first 'cause you called this last week. Didn't name the company, but you called the catastrophe.

   4. JLJason Lemkin49:44

      Well, look, I, I, I think what people missed on, on Block in particular, I know their, their profitability was way up, but I think their top line has only grown, like, 3%, something like that, right? So this story misses-- Everyone missed this. Everyone got suckered in by the press release that said, uh, profit per share, gross profit was up 27%, okay? And it-- That's true, but you only lead with that when you're not growing, right? When you're growing, you lead with revenue growth, right? So what are you gonna do-- I think, I think this story is simpler than it looks. Jack-- They're growing 3%, and he's given up on retur- This is my point. He has completely given up on returning to growth, which I think most public software companies will get to during the course of this year. They will give up. The CEOs will give up. This is my point. They will give up on returning to growth. They will give up, and he gave up, and what do you-- what's your next play? You gotta get more profitable, right? There's a limit where you, where it's not worth-- Like, more than 50% operating margins isn't worth it unless you're an IP licensing shop, right? But it's, it's the get it done because I gotta get to my next state, which is hyper profitable. And so my only point is people reacted to this like this was a high-growth tech company. It has been many years, or at least years, since it's been a high-growth tech company. 3% is pretty tou- I mean, a SaaS company at least can raise prices 4%. At least, at least we have that in SaaS. We just go out to our customers, "Due to the incredible features we've added this year, including changing the colors of our buttons, we're raising prices 6% this year."

   5. RORory O’Driscoll51:06

      Totally. So I, I think there's a lot in that, and I-- First of all, I, big picture agree. A mid-20th billion dollar revenue company, 10% growth a year ago, down to almost no growth last year. You've given up on the revenue growth, so there's only one button to press. You're exactly right, if you wanna make the stock go up. So a couple of comments, though. One is on employee count. You're right, it had ballooned up. You know, there'd been a couple of years where it had grown 50%, and you've gotta calibrate that 'cause interestingly, you know, if you look at revenue per employee or gross margin per employee, compared to SaaS companies, it actually wasn't that far off. But of course, it's not a SaaS company. It's a financial services company, and when you look at it on that basis, where obviously revenue, you know, you, you have all these interchange and pass-through costs, you have-- Relative to best-in-class financial services companies, there wa- there was clearly fat to cut there, and that's why-- and Jason knew exactly why. That's why they did it. But talking about AI, you never use the AI word, and there's two separate uses of the word AI, right? Are you talking about AI on your top line or just AI on your opex? And it's worth pointing out here, this is not an AI top-line story, right? Salesforce is an AI top-line story. AI is either gonna increase your top line if you build, deliver Agentforce, or it's gonna reduce your top line if someone else delivers Agentforce and you don't. It's an AI top-line story, right? This is not an AI top-line story. I'm not gonna use more. At the margin, maybe a little. I'm not gonna say no, but, 'cause Jason will correct me like he did entirely correctly on, um, YouTube. Um, but it's Cash App at the margin or BN, buy now, pay later or Square. AI is not going to fundamentally change the top-line offering here, so this is not an AI positive or negative transformation story. All they're saying when they use the word AI is they're saying, "Dude, we don't make or sell AI to our customers, but maybe we can use AI to cut some opex," right? Is that true or not? I doubt it's true at the 40% level. So when you look at it at that basis, it's not a growth story. It's not an AI growth story. It's a, "We need to make the profits go up and therefore cut all these expenses and plan/hope that some combination of people working harder and AI can allow me to service the customers at a much lower headcount." Entirely, yeah, and get the stock up 20%. The other thing, Jason, you're right, is it is implicitly an abandonment of revenue growth as the plan out of here, at least in the short term.

   6. JLJason Lemkin53:34

      Yeah, I'm not saying that this doesn't give folks cover to make similar cuts, and I, and I, and I'm also not saying AI is exaggerated as the root cause. But, um, but this isn't quite the play people think it is. When growth is decelerated to 10 to three, I mean, it's, uh, you're in a tough spot. And, um, and, uh, I just think where it may inform other CEOs is folks in the, in the, in the, the mid to low teens may, may adopt this. That they may throw in the towel not at 3% growth, but at 12 or 13% growth if there's no accel- If they, if by the end of this year they're like, "You know what? This AI is such a force of nature, but I haven't found a path to acceleration by the end of this year," you may see people copy it just for that reason because they're out of, they're out of ideas.

   7. RORory O’Driscoll54:19

      And I wanna ask a question to you, Jason. I'm gonna-

   8. JLJason Lemkin54:21

      Yeah

   9. RORory O’Driscoll54:21

      ... deliberately do it in a way to separate two things, AI on the top line and AI on opex. Let's assume there's no AI impact on the top line. It's some financial services business that, um, yeah, just has no AI revenue left, right?

   10. JLJason Lemkin54:37

       Yeah.

   11. RORory O’Driscoll54:38

       And let's assume they're efficient beforehand, which I think Block wasn't, right? You could definitely argue there was some effici- What's your mental model of, if I have 100 people doing X this, doing whatever revenue this year, how many per year do you think a well-run company should be trimming based on AI? Do you have a mental model for that? 'Cause I doubt it's 40 in one year. What, what's your mental model?

   12. JLJason Lemkin54:59

       Well, look, just one thing first, then I'll answer the, the question directly. I do think that even if you don't think that a company like Block is threatened by AI, um, the rea- I'm not saying it's like YouTube, but here's the thing I know even just from my own portfolio, folks are building agentic applications that do what Block and others do, and therefore are displacing them because that's where the energy is. There's a lot of products at Block, right? There's a lot of products. But if you go back to some of the more simplistic products like Square, right? Um, as agentic products process point of sale and other transactions and add more value, you d- you can lose, even if you say the atomic version of my product doesn't really need AI. The, the, the, the, it, it, it doesn't, but when these agents are doing everything in your industry... So I t- I think a- the a- agents are capturing so much value that they are... There is just limited budget in every space, and they're gonna take budget away from you. So I don't think... Even folks that are immune I don't think are immune, because the new entrances will find ways to tap into that budget and add more value. In terms of cutting the team, honestly, Rory, I think everyone thinks 40% of their team they don't need today. Every single CEO I talk to doesn't think they need 40% of their team, and so I think, I think that the, the, the Block number will become the default. "I don't need 40%. It's, it's a gift."

   13. HSHarry Stebbings56:19

       I got in a bit of trouble, but I tweeted over the weekend. I, 'cause I spoke to three CEOs between 500 and 1,000 people, and all three said to me that they were cutting by a minimum of 20%. And I think this does create more acceptability around layoffs.

   14. RORory O’Driscoll56:32

       Yeah, it, it, it's the Overton window argument.

   15. HSHarry Stebbings56:35

       Yeah.

   16. RORory O’Driscoll56:36

       I was stunned to learn-

   17. HSHarry Stebbings56:37

       Now's the time.

   18. RORory O’Driscoll56:37

       Yeah. I was stunned to learn that in recent memory, this was the largest percentage headcount cut. Like, I mean, I saw a bunch of data on that where, you know, there's a few 15 or 20% cuts that were bigger, 'cause when someone like Amazon cuts 15 or 20%, it can be 30 or 40,000 people, and this was, I think, from 10,000 to 6,000. No, from 14 to 10,000, I think. Uh-

   19. JLJason Lemkin56:59

       No, it was 10,000 to 6,000.

   20. RORory O’Driscoll57:00

       10,000 to 6,000. You're exactly right. I've been looking at the before and after headcount numbers. 'Cause obviously, look, in venture-backed deals, there are occasional horrible moments when you cut by 50 or 60%, but that's typically when you have 40 people and you lose product market fit. You know, this is the biggest percentage change for a publicly held tech company in the last 20 years, which is just an astonishing certificate. Uh, s- statistic. So I think, Jason, to your point, it definitely expands the Overton window of what's doable if it works.

   21. JLJason Lemkin57:31

       Well, let me give you, just give me a, just a simple example for fun. So I was helping, uh, an, uh, a h- a very hot AI company think about growing their GTM team. We get called into this stuff all the time now, right? And they're going from 10 folks in sales last year to 250 this year. Okay? Let me just ask you, how many of them you think are gonna be great? Great.

   22. HSHarry Stebbings57:51

       20.

   23. JLJason Lemkin57:51

       20 out of the 250? Yeah, maybe. Maybe a few more.

   24. HSHarry Stebbings57:55

       20. 30%. I agree with 50.

   25. JLJason Lemkin57:55

       So look, a few years down the road, if you have to cut, this is gonna be so easy. I mean, I, I know the human impacts are horrible, but any... I, I can't think of any s- company I've ever worked with north of 500 people where 80% of the people were great. It's just managers have manag- It doesn't matter in the age of AI. Now, what's changing is companies are p- gonna be permanently smaller to start. That is utterly change... And that's why, in some ways, Block d- is, is a window to the past. It's like looking out to the s- the galaxies out in the sky, because I don't think we'll build, for the most part, companies like Block the same way today. We may stumble into them in times of plenty, like this guy, th- them going from 10 to 250 in sales in one year. But, um, when you get that bloated, I know we all want to say we don't lower the bar, but managers of managers always lower the bar. First-time managers always lower the bar. It always becomes headcount-driven and hegemony-driven, and 40% of these people, we just have no... Even in AI companies, they have no idea what they're doing. They don't even know what my team... Uh, 500, 800 people? What, what, what are they, what are the 28 people doing in marketing again? They don't know.

   26. RORory O’Driscoll59:00

       Yeah. It, it, it, it, it's funny, 'cause whenever you are, like, this extr- I always wanna argue with you because it's, you're so extreme on it, but, you know, sometimes I find myself realizing that you're correct, right? I mean, 'cause I was just reflecting on this, like, you know, going back to capital versus labor, you know, in the 1950s, 1960s, the biggest companies were, you know, hu- 300,000, 400,000 employees, like General Motors, like Walmart had a mil- billion at one point in time. You know, the amazing thing now is, you know, Apple has 160,000. You know, A- Amazon... Sorry, Google has 190,000. The trend says you're right, because even before AI, the valuable companies have been the high IP, low headcount people. And you look at the trend today, you look at the revenue per employee at someone like a Cursor, it's, you know, it's about the same already as Apple, maybe even a little higher than Apple. Apple's about 2.6 million revenue per employee. Um, you know, Google is 2.1. So these guys are already at that level at a much smaller scale. So I don't know if it's as extreme as you say, but the trend has definitely proven you're right, which is more and more valuable companies with less and less people. I don't think it trends to one. I don't think it trends to the billion-dollar single-person company, but the direction of travel is on your side of the table.

   27. HSHarry Stebbings1:00:13

       I do think we'll see an acceleration of layoffs, though, with founders appreciating this kind of acceptability of doing large layoffs in the wake of this. That's what I hear is happening.

   28. RORory O’Driscoll1:00:23

       I always hate the word acceptable. I think the more common, the harsher comment is necessity. You know, in other words, acceptability sounds like it's a social thing. I wouldn't like to get rid of people's career just 'cause it suddenly, quote, "became acceptable." If you are subscale, growing sub-10 and kicking off 10%, if you don't take the actions and someone e- I... There's a... Remember, price lists all markets, as I repeat myself over and over again. If you're only doing 10%, less than 10% revenue growth and 10% free cash flow, you'll be trading at four or five times and three or four times perhaps, and someone will know how to come in and buy you and make those changes, and then a month later they'll be worth eight times.

   29. JLJason Lemkin1:01:03

       I think you do have a choice. I think Harry's point's an important one, and I don't even think it even ma- matters at some level because it's the past, but I think the Block thing frees CEOs to make what they believe is the right decision for the company without overanalyzingThe horrific human impacts. They, they are now free to sit in the board meeting next week probably. These meetings will probably all happen in March, and they'll be like, "Guys, let's just talk about Block calmly. I'm not saying we should do this, but what is-- Given that we missed the quarter again for the fourth quarter in a row, guys, we missed... You said we were coming back, and I get it, but, but we're at $110 million in revenue we missed or whatever, to $400M, we missed the fourth quarter in a row. What could we do like Block?" It's at least gonna be a conversation, right?

   30. HSHarry Stebbings1:01:44

       I think also the extent allows them to cut deeper. The 40% is a big number, and I think it allows them to go-
9. 1:02:45 – 1:17:01

   Cursor Hits $2BN in ARR… so not Dead?

   1. HSHarry Stebbings1:02:45

      Yeah. Why do, why don't we, why don't we talk about one of the future companies that we mentioned on the-

   2. JLJason Lemkin1:02:49

      Yeah

   3. HSHarry Stebbings1:02:49

      ... revenue head count? 'Cause I'm confused, chaps. I speak to everyone about their product teams and their engineering teams, and everyone has moved off Cursor and moved to Claude Code. I tweeted about this the other day, and thousands of people comment this. And then, and, and everyone is in universal, "Cursor's dead. Claude Code rules all." And then Cursor announces their movement from a billion to two billion [chuckles] in revenue in, in three months. Um, rumors of secondary round being done at 50 billion. Um, maybe in my favorite saying that Rory often said about benchmark, "Reports of my death have been greatly exaggerated," and I thought of that here. What the...

   4. JLJason Lemkin1:03:36

      Well, just one comment. Maybe Rory has, uh, deeper insights second. I will say it's, it's funny because, um, this is, this is one of those situations clearly where VC is judging the world based on their portfolio versus the real world are out of sync. Okay? Literally, right before this all came out and blew up, I had two board meetings of my fastest growing portfolio companies. One of... And they all joked about how folks weren't using Cursor. And one said, um, "B- but mostly it's only grandpas at the company that still use it." And the CTO stood up, who's pretty young, and said, "Grandpa over here. I still enjoy it." And he's like, "There's a couple of us that still use Cursor." And then I was with another company that has blown up all of their, uh, AI... And like, and, you know, there were two people that used Cursor. So you, you heard these stories all the time, and you just assumed that every- everyone had mainly moved to Claude Code directly, right? And product teams said too, and you just assumed that's the way the rest of the world works. And that's why Cursor was smart to r- to, to, to slip this, uh [chuckles] this news release to Bloomberg that they'd, they'd happened to double in the last 90 days. So our portfolios are not representative. I think the an- listen, this is much as I know, other than we all made the same jump to the conclusion based on our 40 portfolio companies. But if six- you know, they said 60% of our revenue is enterprise, and probably the answer is right there, right? Is that this is a, this is a trusted tool with a lot of things, including guardrails that can n- that can manage agentic swarms and others. Um, they're pointing out how banks use them and other conservative enterprises and are just rolling them out. And I don't know this to be true, but if Cursor's the more conservative choice, that's not the worst place in the world to be, especially if you can pick your model. M- maybe that's the answer, but it's not, it's not my five fastest growing investments. That's not where they're crushing it today versus six months ago. Um, but it's so funny we got it wrong. Like, numbers don't lie. [sniffs]

   5. HSHarry Stebbings1:05:26

      The feedback that I got from the tweet was v- was very simple from insiders. It was twofold. It was, one, that you have enterprise crushing it, exactly as you said. And then two, actually, just annual subscriptions and their cycles not being complete yet. And so there will be a wave of unsubscriptions, and that churn will be real on the consumer side-

   6. JLJason Lemkin1:05:45

      But that can-

   7. HSHarry Stebbings1:05:46

      ... but they are annual

   8. JLJason Lemkin1:05:46

      Rory's the expert here, but that, that, that-- even if that's true, and it's a really interesting topic in general, w- w- we, we sh- maybe it's worth talking about. You don't go from one to two billion in 90 days, uh, i- if you're having massive chu- like deferred churn at the end of the contract. That's a different issue that's gonna come down the road. I mean, that is a, that is growth that is just ju- massive market pull. Like epic market pull to grow that quickly, right? Just to sign the deals takes so much energy. [laughs]

   9. RORory O’Driscoll1:06:14

      And, a- a- yeah, that's exactly... And I think the lesson is never underestimate big markets and momentum, right? You know, y- y- you know, if you come into last year with the momentum they had or the middle of last year, even if, you know, things are slowing at the margin, even if at the marginal startup you're losing to Claude Code, you know, you've built your name, you've built your brand. Every enterprise is trying to adopt coding. The adoption cycle there isn't, "Oh, I tried, um, Cursor yesterday. I'm gonna try Claude Code tomorrow." It's like corporate's got to approve, purchase order's got to be raised. We got to have security review it. Legal's reviewed the contract. We signed a deal. You're not gonna run in next day and say, "We should switch to Claude now." It's like, "No, we're gonna use this for the year. We'll do an eval mid-year, and we'll think about it next year." I mean, Jason knows better than me. That's-- So I-- The bigger high here is when you're in a great big ma- market with a massive trend of adoption in the course of, you know, two, three years, if you've got your share of that, you know, y- you're, you are, you're gonna do just fine. I mean, you know, at the margin, are you probably losing share?Well off the Claude, at least in the last three months since the last set of models probably, but it still means a big outcome. I mean, you know, we talked about this two or three months back. It's clear that they're gonna be one of the two or three players at scale, and scale gives you a lot here. I would add, you're right, Harry, numbers don't lie, but I haven't seen the numbers, right? And I haven't seen the revenue, and I haven't seen the gross margin. But, you know, you can easily see how someone with an early lead would be able to maintain that lead even when, you know, you've got a super credible competitor who can bundle a product into an adjacency, taking some of that space, because the market's expanding enough to cover a multitude of sins. The knife fight... Maybe this is the real sense, Harry. The knife fight doesn't start until the TAM is like 60, 70% saturated, which by the way, happened to SaaS in 2021, and that's when the knife fight started, right? Suddenly you're like, you know, one in two of every new customer is a switch. Now it's time to kill the other guy. Until then, let's all grow one big happy family. We all have a win rate of 60%. We don't compete against each other. All those happy things that you hear in the board when the market is so big they we're not overlapping. Once things get saturated, then it gets uglier.

   10. JLJason Lemkin1:08:31

       Yeah, there was, I think there was a twee- I think there was a quote or something from someone at Cursor. I might be wrong who said it, but I think it fits, it's an interesting point, which is that, uh, Barclays had just rolled them out, okay? That was the first agentic coding product that was approved at Barclays, okay? And banks are a big deal because they have the biggest budgets, but they are conservative at the same time. It's an odd dichotomy, right? They're super aggressive and super conservative. And, you know, m- we might have gotten Cursor wrong. I mean, Cursor is more enterprise. You can run it without data retention. You have full SSO, you have role-based access controls, and all these things that you just got to build in the enterprise, the right audit logs, the right hooks. And to a, to a CISO or CIO, as long as the product works, that could be 10 times more important than what someone in the other building wants, okay? And I think as-- And the other thing that's happening right now, good God, is, you know, it's been in, in all of these products for a while, but in Claude 4 Opus, these swarms of agents got really good, okay? These swarms of autonomous agents. And Michael from Cursor wrote this. Not only did they leak the $2BN, but he wrote a post which, uh, you know, a tweet article, which seemed kind of weird, but talking about how the whole future of Cursor is sort of safely managing these autonomous agents, not, you know, the early days were, were clicking tab to finish a, to finish a line of code, which was what Codium and Windsurf got going with. And if the-- And, and when you have swarms of autonomous agents, good God, they're not totally safe. Not-- Much as we all love our open claws, they're not totally safe. And the more of them you have that are autonomous and the more they can touch your data, th- they're, they're not safe. And, and, you know-- And so if Cursor can do both, if it can unlock the power of swarms of autonomous agents running in real time, running, uh, concurrently and make it safer, I would got to think 95% of CISOs want that one. Yeah, and, and, and, and a lot of conservative organizations, the cranky CTO doesn't get to make all the decisions like at our startups. You know, you-- Uh, uh, at our startups, the CTO decides, you know, what the team's gonna use, right? Uh, but good God, at a bank, what if you repeatedly leak everybody's confidential data? I mean, it's exhausting to have to go through that. Th- this is not a joke. Like, every agent will leak data if it's allowed to. Like, every single agent will. That's how goal-seeking works. Being the most en- Like, being the most enterprise leader in these mature spaces, you can, you can win because 99% of the world is not like us.

   11. RORory O’Driscoll1:10:57

       No, th- th- g- By the way, that, that's the part-

   12. JLJason Lemkin1:10:59

       Barclays is not like us.

   13. RORory O’Driscoll1:11:01

       Jason, sold. I mean, all you have to do is look at Azure's market share in cloud-

   14. JLJason Lemkin1:11:06

       Yeah

   15. RORory O’Driscoll1:11:06

       ... where for 20 years, the value proposition, having said in millions of board m- not millions, but tens of board meetings where we've talked about, you know, we're running on AWS, should we run on Azure, our customers want to run on Azure. And the value proposition of Azure for 20 years was it's not nearly as-- i- it's not as good, but it's getting there, and Microsoft, um, have great relationships with every CTO, so it'll be fine. You're exactly right. You know, there's a large amount of s- software selling that can be done on the basis of safety and integration and selling all those boring bits. What you're dealing with is a massive market where everyone's making adoption decisions in an extraordinarily short period of time. And let me repeat what I said earlier, is they're both gonna crank as fast as they can. They're gonna... And as the mark-- as people make initial decisions and more and more of that market has an initial buy decision made, you're gonna start running into each other. Is that gonna be later? You know, the, the competition's just gonna intensify. They're not... So, I mean, let's just do it. If Cursor went from one to two, I think in the same period this year, I saw some Anthropic statement about, you know, we've added, what, $2BN in ARR year to date. So both of them are exploding. This is a huge... This is the biggest freaking market, and those are the two most viable players in it.

   16. JLJason Lemkin1:12:21

       Look, it's, it's-- Going back to last week, it's the Fortnite effect. Like, Claude will-- code will keep doing more, and Cursor has to keep ahead. That's the job, right? I, uh, uh, you know, uh, apparently 60 to 70% of Cursor model calls are still using Anthropic's API, okay? So even of that $2BN, a lot is flowing back through to, to mom and dad. This is the point Michael Cannon-Brookes made the other day. And, and Claude, even if they... They may not even care about Cursor as a team, but they will inherently build all these features as an enterprise play, and so that, so... And so the bubble will shrink, and then Cursor just has to do more. But, and what we've learned this week is they've kept ahead. Th- they have kept ahead of the Fortnite storm. They've kept... And every, every CEO out there crying and, and saying how hard it is, uh, and that hasn't kept ahead, they're gonna be bi- that Fortnite storm's gonna shrink to a pixel, and they're gonna die.

   17. RORory O’Driscoll1:13:07

       Jason, I think you're right. That is the soundbite. I mean, the, the correct response there is, congratulations. In the grace-- In the great race for what looks like a $50 to $100BN minimum coding prize, you have gr- you know, you have conquered level three, which means unlike a whole bunch of people who are stuck in a back-end level two, you now get to go to level four and play again, right? It turns out it's hard to make $100BN, but you deserve, they deserve all the credit for, you know, being the early winner, cranking, and keeping revving the story. I mean, it also, just again, back to the swarms of agent, it's just worth stating, and again, this is obvious, but just to say itHow... If you look at the, the Cursor story, I mean, he spelled it out in that kind of post, which I thought was very good. If you're not... This is back to Jason's point. If your story is not changing significantly every six to nine months, you are probably falling behind. In fact, not probably, you are falling behind. You know, you were autocomplete, then you were an IDE, now you're agents, now you're swarms of agents. These models are moving so quickly, that the technology is moving so quickly, that if you think you can rest on... Every, every six months you get, you get to re- The, the prize for winning is to reinvent the company from scratch and the product from scratch every six to nine months. Congratulations, it's a fun game. But the prizes are great.

   18. JLJason Lemkin1:14:21

       You know, the fun- the one thing I'd still love to know about Cursor, and it is incredible the growth, is, and, and, and not to be very 2025 about it, but the Claude API is expensive. You know? I've, I've measured our API calls. It's so much cheaper to run something inside of Claude. My God. So, like, I mean, I'm on Claude Max. I, I... Uh, Emilia runs out of credits using Claude Desktop and everything, or she does, she runs out of time. She doesn't even run out of credits. But what I get for 100 bucks a month, and then I track what some of our API calls cost, and, and for, for, for a good prompt, a good call, it can be well over a dollar. Okay? And maybe for code it could be more. Like, that adds up really quickly, and it is, whether you think it is subsidized, whether you think Anthropic is subsidizing its own native use or not, it d- their margins are strong. It doesn't really matter. So that is still the more, like, hooray, you hit the top line, but when you're still paying probably rack... Like, I can't imagine Anthropic is so desperate for Cursor's business they're giving them a massive discount. Right? Harry may find that out through his network, his in-per... But if they're paying close to rack rates, which I think they are, because they also say you can bring your, your, your key with you. Like, bring... If it's cheaper, bring your own API key with you. It says there's not much of a discount. That remains your, your at Anthropic's whim because they're already subsidizing their own product.

   19. RORory O’Driscoll1:15:40

       Totally.

   20. JLJason Lemkin1:15:41

       W- w- we're lucky people are not that pri- no matter what anybody says, we're not that price sensitive today.

   21. RORory O’Driscoll1:15:46

       Yeah. It was super interesting. I got to give a shout-out to Tomas, who, who we had on here a few months back. Did an interesting post just today. As you know, basically, would you buy... I can't remember. It was a good metaphor. I should remember and I can't. Basically, just looking at the pricing of, you know, top of the line frontier models versus six-month-old open source Chinese, um, open source models, and you know, the 10, 15, 20x difference in cost per token. It was a good piece. And you're right. Right now people are paying full rack rate for the new stuff because the new stuff is so great. But if the pace of in- I mean, his point he was making is it's only six months from so new that it's amazing to, oh my God, you can get it open source for 120 at the price. And it just really, again, brings home pace of change and the need to be on top of it.

   22. JLJason Lemkin1:16:31

       I think OpenClaude confused things, but the power of running swarms of agents autonomously to build software, the rate of change, we think the rate of change is, like, I know you didn't literally mean it, but the, the way you described Tomas's thing suggested change might be slowing down, right? And we might get a benefit of these cheaper models. It just... I know you didn't suggest it, but it, there was a hint of it. I think i- it's going to radically accelerate this year. Radically acc-

   23. HSHarry Stebbings1:16:57

       Can you help me understand why, Jason? 'Cause I actually don't un- Like, help me see what-

   24. JLJason Lemkin1:17:00

       Because what changed in

10. 1:17:01 – 1:25:15

    How to Pick Winners in AI?

    1. JLJason Lemkin1:17:01

       December, what ch- So, like, listen, I've been on Replit, not to talk about forever about Replit and Lovable, but I've been on it since June, and when I started, you couldn't finish anything. You'd ask Replit or Lovable to build an app. You remember in the early days, Harry, and you'd get a lot of buttons that didn't work. Okay, nothing worked. They'd be fake buttons. Everything was fake, right? And then to O- then it got better and better, and then by December, especially with Opus 4.5, all of a sudden it could fit... It wasn't just you could have a code for hours, whatever. That doesn't really matter for a lot of use cases. It could act, these... It could finish apps that worked. Like, really worked, right? And then on our team, Emilia took it over and she built an entire AI VP of Marketing that is better than anything on the market, and she finished it without, like, any drama. Like, it took work, but she just completed it. It's in production every day doing all of our marketing, not just execution, but it has all the ideas. It's like Hal. It wakes up every morning and tells us what to do with data. It's a little freaky. And, and one of the reasons was, one of the many reasons was these multiple agents running in swarm and working on the software together. An architect, a security expert, a backend expert, a database expert, and Re- Replit k- just did one thing early, which is the... And, and maybe Cursor did this too. They built a framework so you could use these multiple agents together efficiently before you could in Claude, right? You could do it earlier. And so it, it, it, now that someone that's just very smart can birth an AI VPM that m- marketing that literally works, in six months, the applications we're able to build, our jaws will just drop. We'll just, we'll just drop, and already I don't even know what to do with demo day. We literally got a sponsor, uh, that, that, that ju- that just raised 50 million. Their, their website was all built in Lovable. We just, we were just laughing about it, but maybe that's not funny anymore. Maybe, [laughs] maybe these apps are so good.

    2. RORory O’Driscoll1:18:45

       What do you mean you don't know what to do for demo day? I think I, I think I get a bit here.

    3. JLJason Lemkin1:18:48

       Be- because you can... Because if it's... You know, uh, um, what's his face from AngelList said software is u- uninvestable. It should've been on the show. It should've been a topic. Okay? I agree, and I, and I'm struggling with it, and, and, and what I mean is you can show up to a demo day and build anything you fucking want now in, in, in, in, in, in any of these coding tools that's really good, that is so much better than anyone would show up to a demo day 24 months ago. 24 months ago, half the folks would show up, they didn't have a product. It was barely working. You could make fun of it. Um, it, we went through it. And now you go, and, like, there's no excuse to not have something that's, like, great. Hey, I mean, w- I could take what we built. Emilia and I could show up to YC and say, "We built this AI VP of Marketing. It's managed a million marketing transactions." Sh- it would blow your mind. Like, I'm raising at 80 pre. Po- post. Po- post.

    4. RORory O’Driscoll1:19:38

       Just to be precise, Jason, I mean, 'cause the sentenceSoftware's uninvestable. I mean, if software's uninvestable, why bother making the demo at all? Because it's not investable. So clearly saying is you don't invest on a demo anymore 'cause a demo simply means 24 or 48 hours earlier you did, you kicked off, um, a coding project and you got it done. So there's no information in the demo anymore, right?

    5. JLJason Lemkin1:20:01

       I don't think that-- I think for most cases, I don't think there's any signal or information in a demo anymore. Yeah.

    6. RORory O’Driscoll1:20:06

       Agreed. So I think at the early stages there's no... That's, that's, so that's what you mean by demo, yeah, 'cause the software is investable. Obviously, it turns out Microsoft stock is holding up pretty well. So software is invest- is uninvestable. It's one of those hyperbole statements, right?

    7. JLJason Lemkin1:20:19

       Well, what I mean is, listen, I had-- One of the reasons I was modestly successful when I started investing is I was a founder that turned into an investor, okay? And one of the superpowers you get when you make that transition is you know what's better than you. You know the CEOs that are better than you, and you also know who's better at building software. So when I would meet a founder that, uh, week for week, pound for pound, had shipped better software than I did, um, I knew they had a decent chance. Like, if the demo blew me away and the CTO blew me away, even if they're, even if they had a couple thousand in revenue, I'm like, "These kids can't lose. They, they can't lose."

    8. HSHarry Stebbings1:20:54

       Jason-

    9. JLJason Lemkin1:20:54

       Yeah

    10. HSHarry Stebbings1:20:54

        ... can you name one founder, one founder who is able to create, architect, maintain agents in the way that you have done? And I'm not blowing smoke up your arse, but Rory, I think this is where Jason's like, "Oh, it's gonna accelerate so fast and so, so fast." I don't know one founder who has built an AI VP of marketing in the way that Jason has done.

    11. JLJason Lemkin1:21:16

        I do think I'm ahead of most of the... Listen, there's obviously hyper a-- There are folks just building agents that are so far ahead of me, I can't even-- They, they mock me on X, and I deserve it. I, I don't take... You know, if you're better... But I would say for your f- for your average AI-focused B2B founder, I am ahead of them on this stuff, right? 'Cause I've been doing it for, for six months longer. They'll eclipse me later in this year, but, but, but, but they don't get it. They're, they're most... You-- I'm sure you're getting investor updates in March which even talk about how they're stunned with how much more productive their team is, right? We're all getting these in, like they're just, they're just stunned. Like I know we've been talking about this, but now it's finally hap-- Like it's not just a few more pull requests and this and that. They're literally, their jaws are starting to drop in March because of what they built in February and late January with these tools now, because they can finish stuff. Like they can finish stuff. And when that 10X's this year, here's the thing, like your, the rate at which your product has to accelerate will be like nothing we've ever seen before. Nothing we've ever seen before, right?

    12. RORory O’Driscoll1:22:18

        That's the real version of software is not investable. The real version is there's going to be infinitely more software, so your ability to carve out this little discreet... So software alone will not be a competitive advantage. So there will still be extremely large demand for software, and there as, as a result of that will be very big software companies, but they will have to do something more than, "Hey, I wrote a bit of code for this particular vertical and this particular use case." That's not what's gonna get you there. It's gonna be s- it, it's gonna be some other elements of competitive advantage around network, around distribution, around moat, around vertical knowledge. I mean, I think that's the, that's the true version of the statement.

    13. JLJason Lemkin1:23:00

        I think there's... Yeah. I think it may... And then listen, at the later, you know, investing in Eleven Labs today is, y- y- you know, the, the, it's a very coherent bet, right? It just makes this idea that you could judge, or at least if you had the right background, that you could judge early stage experience better than many and write a check very dif- and even more so, listen, I had two investments. They weren't huge investments, but they weren't tiny. Recently, they did a million or more their first week in market, AI investments, because the demand was so strong. It's not... But, but let me... One of them, literally 45 days before, 50 days before they were showing me a demo, I'm like, "Guys, I love you, but [laughs] this... I know all the vendors in the market. I like, I, I put it back in the oven for a little while. I don't think it's competitive." And then, and, and, and, and then it, you know, you roll out and you dominate the market. There's a lot of reasons, but the, that level of pace we just didn't see before and it's gonna accelerate, so how the hell do you pick favorites? How the hell do you pick favorites unless... They had, what's his name, Frankel from, um, what's his name, Harry, that did Sono-

    14. HSHarry Stebbings1:24:04

        David Frankel. Yeah.

    15. JLJason Lemkin1:24:04

        Yeah, he was the... He said he, he, he had as one of his never to braggy ones. I was the only pre-seed investor in, in Sono, which is just 300 million in two years, right? Um, and he's like, "Why did I invest in it?" He's like, "Well, PhD from this, master's in that, understood the whole space. I met him. Everyone said he was the smartest guy in the world." I mean, that signal still exists in 2026. [laughs]

    16. HSHarry Stebbings1:24:24

        Do you, do you, do you wanna hear a joke?

    17. JLJason Lemkin1:24:26

        What's the joke?

    18. HSHarry Stebbings1:24:27

        He tried to bring me into that round.

    19. JLJason Lemkin1:24:29

        Well, you should've done it. You should've done it.

    20. HSHarry Stebbings1:24:32

        Oh, no. Boys-

    21. JLJason Lemkin1:24:34

        He probably saw that, that you couldn't see quickly though, that, that's a, you know, that, that level of depth, right? That you couldn't see, right?

    22. HSHarry Stebbings1:24:41

        Rory, are you gonna have your bets ready for next week?

    23. RORory O’Driscoll1:24:44

        I will have them for next week, Harry. I will have them.

    24. HSHarry Stebbings1:24:47

        That's fair. We get... I'll give you a, an allowance 'cause it's 10 o'clock in London, and, uh-

    25. RORory O’Driscoll1:24:51

        No, I, I appreciate that though. I, and I will. I genuinely will. But I'm still up on my world cloud. I'm up about six, I think six or seven. I was up 6% at one point in time. I'm not s- which is the wisdom tree index of all SaaS. I just... Yes, as the idiot version of that. So worth remembering. Now, I don't, I haven't looked at today and it's been pretty grim, so who knows? But yes, I will have my act together for next week because it is late in the day there in London.

Episode duration: 1:25:25